cd /news/artificial-intelligence/show-hn-indicate-transliterate-indic… · home topics artificial-intelligence article
[ARTICLE · art-119739] src=github.com ↗ pub= topic=artificial-intelligence verified=true sentiment=· neutral

Show HN: Indicate: Transliterate Indic Languages with PyTorch and LLMs

Indicate, a new open-source tool for transliterating 12+ Indic languages to and from English, was released on Hacker News, offering both a PyTorch-based local model and LLM backends with auto-detection of source scripts. The tool supports bidirectional transliteration, batch processing, and structured JSON output, with Python 3.13+ required and weights downloaded from Hugging Face on first use.

read9 min views1 publishedSep 3, 2026
Show HN: Indicate: Transliterate Indic Languages with PyTorch and LLMs
Image: Michielbdejong (auto-discovered)

Indicate provides high-quality transliteration between Indic languages and English using both a traditional PyTorch model and state-of-the-art LLMs (Large Language Models).

🔀 Composable Backends: Chain a word table, a local model and an LLM in any order🌍 Multi-Language: 12+ Indic languages, with the source script auto-detected🔄 Bidirectional: Supports both Indic→English and English→Indic transliteration🛡️ Production Ready: Safe file handling, atomic writes, backup support📊 Structured Output: Rich JSON format with metadata and error handling⚡ Batch Processing: Efficient processing of large files with progress tracking

Hindi • Tamil • Telugu • Bengali • Gujarati • Kannada • Malayalam • Punjabi • Marathi • Odia • Urdu • Sanskrit ↔ English

We strongly recommend installing indicate

inside a Python virtual environment (see venv documentation)

Requirements: Python 3.13+

pip install indicate
pip install indicate

export OPENAI_API_KEY=your-key
export ANTHROPIC_API_KEY=your-key  
export GOOGLE_API_KEY=your-key
pip install indicate

The Bengali word table downloads from the pinned model-assets repository on first use and is then cached. It is compiled from a shared, LLM-labeled electoral-name corpus into one deterministic native-to-Latin lookup; the multi-million-row source CSV is not duplicated in this repository or package.

Hindi and Punjabi tables are different: they derive from data/hindi.csv.gz

(which blends CC-BY-NC IIT Bombay pairs) and data/punjabi.csv.gz

(from a restricted electoral-roll deposit), neither of which is ours to redistribute under MIT. Build those from a checkout:

export INDICATE_DATA_DIR=~/.local/share/indicate     # where your tables live
uv run --group train python training/build_lookup.py --lang hindi
uv run --group train python training/build_lookup.py --lang punjabi

INDICATE_DATA_DIR

is where the builder writes and where an installed package looks first. Without it the table lands inside the checkout, which a pip install

ed copy in site-packages

will never read. Keep it exported and indicate languages

flips that row from unavailable

to ready

:

Direction                 Backend   Status
bengali -> english        lookup    downloads on first use
                          llm       needs an API key
punjabi -> english        lookup    ready
                          model     ready

Without a Hindi or Punjabi table nothing breaks: lookup

declines every word and model

answers them. Bengali is lookup-only locally, so an unavailable table is reported as an error instead of silently returning blank text.

One command, one function. The language and the backend are arguments, not separate entry points.

indicate transliterate "राजशेखर चिंतालपति"

indicate transliterate "ਰਵਿ ਸ਼ਰਮਾ"

indicate transliterate "বৰুৱা"

indicate transliterate "नमस्ते" --from marathi --engine llm

indicate transliterate --input names.txt --output roman.txt --format json --backup
indicate transliterate --input names.txt --output roman.txt --dry-run

indicate languages

indicate info

python -m indicate

does the same as the indicate

script, for when the console script is not on PATH

.

import indicate

indicate.transliterate("राजशेखर चिंतालपति")  # "rajshekhar chintalpati"
indicate.transliterate("ਰਵਿ", source="punjabi")  # "ravi"
indicate.transliterate("नमस्ते", n=3)  # 3 ranked candidates
indicate.transliterate_batch(["हिंदी", "मुंबई"])  # ["hindi", "mumbai"]

indicate.supported()  # {(source, target): (backends...)}

A word is answered by the first backend that will answer it. The chain is an argument, so you decide how much machinery each word is worth:

chain what it does
lookup, model
default — read the table, decode the rest locally
model
decode everything; what a benchmark must use
lookup
table only, "" on a miss — "is my corpus already covered?"
lookup, llm
the table intercepts the paid path
lookup, model, llm
escalate to a provider only what both decline
llm
ask a provider for everything
indicate transliterate "मुंबई" --engine model
indicate transliterate "मुंबई" --engine lookup,llm --provider openai
indicate.transliterate("मुंबई", engine=["lookup", "llm"])
indicate.transliterate("मुंबई", engine="model")

A backend that cannot serve a direction is skipped; if none remain you get an error naming what would work, rather than a silent fallback onto something that costs money:

$ indicate transliterate "வணக்கம்"
Error: no backend in ['lookup', 'model'] supports tamil->english;
try engine=['llm'] or see indicate.supported()

That is UnsupportedPairError

. A different failure gets its own type, because the two mean opposite things:

  • a backend that declined— it loaded its table and had no entry for that word — is ordinary and silent.engine=["lookup"]

over an uncovered corpus declines everything and returns""

, which is the whole point of asking. - a backend that was unavailable— no table built, no weights, no network — answers nothing because it could not run. Wheneverybackend in the chain is in that state you getBackendsUnavailableError

naming each one and what to do about it, rather than an empty string that looks like an answer.

try:
    indicate.transliterate("राजशेखर")
except indicate.BackendsUnavailableError as exc:
    print(exc)  # nothing could answer 1 word(s): lookup has no table (build ...

Known words are answered from the word table and never reach the decoder. On Punjab electoral-roll text that covers 99.1% of tokens, so the model handles the tail: 42x the end-to-end throughput (10,937 tok/s against 258), and an input that hits entirely never even imports torch, which is worth 4.4x on cold start (0.10s to first answer against 0.44s). training/bench_lookup.py

reproduces both.

It is also more accurate than either component alone, because the builder declines to answer where the training corpus has no majority and lets those words fall through: on the Dakshina test set, 78.8% exact against the model's 76.2% for Hindi, 77.6% against 77.0% for Punjabi.

Two caveats worth knowing before you rely on those numbers. They are measured on electoral-roll names; on general Wikipedia prose the same table covers 56.9% of tokens, not 99.1%, and the cold-start win largely disappears because a sentence almost always contains a miss. And the shipped table contains 908 of the 2,500 Dakshina Hindi test words, so the Hindi accuracy figure is optimistic by an unknown amount. training/build_lookup.py --eval-clean

builds a table with every eval word withheld.

Use --engine model

(or engine=["model"]

) to measure the model by itself — benchmarks must, or they score memorization. training/seam_check.py

checks that mixing table and model output in one string stays stylistically consistent.

For whole-sentence transliteration with context, use the client rather than the engine chain — the chain resolves word by word:

from indicate import IndicLLMTransliterator

transliterator = IndicLLMTransliterator("hindi", "english")
transliterator.transliterate("राजशेखर चिंतालपति")
transliterator.transliterate_batch(["राजेश", "गौरव", "प्रिया"])

For millions of tokens, indicate.batch

submits to a provider's async Batch API with checkpointing, and answers what it can locally first:

from indicate.batch import transliterate_tokens_batched

pairs = transliterate_tokens_batched(
    tokens,
    "punjabi",
    "english",
    checkpoint_path="run.jsonl",
    engine=("lookup", "llm"),  # default; ("lookup","model","llm") goes further
)

--format json

works with every backend, not just the LLM. One line of input in, one entry out, with the chain that answered it recorded per row:

{
  "metadata": {
    "source_language": "hindi",
    "target_language": "english",
    "timestamp": "2026-08-14T07:40:08.697757+00:00",
    "total_lines": 1,
    "successful_lines": 1,
    "failed_lines": 0,
    "format_version": "1.0",
    "encoding": "utf-8",
    "description": "Indic language transliteration results from indicate package"
  },
  "results": [
    {
      "line_number": 1,
      "input_text": "राजेश कुमार",
      "output_text": "rajesh kumar",
      "source_lang": "hindi",
      "target_lang": "english",
      "confidence": "lookup,model",
      "error": null,
      "processing_time": 0.07029390335083008,
      "timestamp": "2026-08-14T07:40:08.697423+00:00"
    }
  ]
}

confidence

holds the engine chain, not a probability — the local model's beam scores are not calibrated, so publishing one would invite a comparison it cannot support.

🔒 Input/Output Validation: Prevents accidental file overwrites⚛️ Atomic Writing: Safe file operations using temporary files💾 Automatic Backups: Optional timestamped backups of existing files👁️ Dry Run Mode: Preview operations before execution

Resumable runs live in indicate.batch

, which checkpoints every resolved token to disk and picks up where it left off.

indicate transliterate "text" --engine llm --provider anthropic --model claude-3-opus

indicate transliterate --input results.json --from english --to hindi --engine llm

indicate transliterate --input names.txt --engine lookup

lookup | model | llm | | |---|---|---|---| Directions | Bengali, Hindi, Punjabi → English | Hindi, Punjabi → English | 12+ languages, any Indic pair | Setup | Bengali downloads; build Hindi/Punjabi | none | API key | Speed | 10,937 tok/s end to end | 258 tok/s | network-bound | Cost | free | free | per API call | Offline | ✅ | ✅ | ❌ | Coverage | only what is in the table | every word | every word | Answers with | the corpus label | a decode | the provider |

Both speeds are end-to-end on roll names, measured back to back on one machine, so the ratio is the meaningful part. The table itself serves 16.9M reads/s once loaded; that number describes the dictionary, not the pipeline, and quoting it as throughput would overstate the win by three orders of magnitude.

indicate languages

prints which of these are available for a direction on your machine.

Clone and install:

git clone https://github.com/in-rolls/indicate.git
cd indicate
uv sync  # or pip install -e .

Run tests:

uv run pytest                       # everything
uv run pytest tests/test_engine.py  # one file

Model weights and lookup tables are gitignored, so a fresh clone skips the tests that need them and prints what is missing with the command that builds it. To make those skips into failures instead — which is what CI does, after building the tables from the committed corpora:

uv run pytest --require-artifacts

Test the backends:

indicate transliterate "हिंदी" --engine lookup,model

export OPENAI_API_KEY=your-key
indicate transliterate "हिंदी" --engine llm

The datasets used to train the model:

Indian Election affidavitsGoogle Dakshina datasetESPN Cric Infofor hindi version of theenglish scorecardIIT Bombay English-Hindi Corpus

The v2 models (trained on our data + the public Aksharantar corpus) are benchmarked against AI4Bharat IndicXlit — the same direction (native→Latin), the same test sets, the same metric (Top-1 exact-match, match-any-reference). Training is leakage-filtered so no eval word appears in it.

Model Dakshina (gold) Held-out-own names¹
Hindi → English 74.4% (IndicXlit 73.2%)
52.8% (IndicXlit 49.7%)
Punjabi → English 71.9% (IndicXlit 73.2%) 56.9% (IndicXlit 53.5%)

¹ Held-out slice of our own electoral/affidavit names — the cleanest comparison, since IndicXlit never trained on it. v2 matches or edges IndicXlit on the gold benchmark and beats it on the deployment domain. Primary metric is Top-1 exact-match; CER (character error rate) is the soft companion. Reproduce with training/eval.py

and training/compare.py

.

Below is the edit-distance distribution on the test set (0 = exact match):

Rajashekar Chintalapati and Gaurav Sood

The project welcomes contributions from everyone! In fact, it depends on it. To maintain this welcoming atmosphere, and to collaborate in a fun and productive way, we expect contributors to the project to abide by the Contributor Code of Conduct.

The package is released under the MIT License.

── more in #artificial-intelligence 4 stories · sorted by recency
wici.ai · · #artificial-intelligence
WiCi One
── more on @indicate 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/show-hn-indicate-tra…] indexed:0 read:9min 2026-09-03 ·